Not so much turned off as simply not present for the execution that generates the checkpoint. Ali
Sent from my ARM powered device On Jun 10, 2012, at 8:54 AM, Ira Ray Jenkins <[email protected]> wrote: > Does this imply the caches can be turned off to checkpoint? And later > turned back on? Or that you cannot checkpoint a system that is using > caches in the classic memory system? > > On Sun, Jun 10, 2012 at 7:42 AM, Nilay Vaish <[email protected]> wrote: >> In the classic memory system, caches are not checkpointed. Therefore, while >> taking a checkpoint, you should only have the physical memory in place. Or >> else, the memory state captured in the checkpoint would not be the actual >> memory state. >> >> -- >> Nilay >> >> >> On Sun, 10 Jun 2012, Ira Ray Jenkins wrote: >> >>> The original command line is as follows: >>> >>> build/ARM/gem5.fast --outdir=test.m5out.fast configs/example/fs.py >>> --caches --l1i_size=32kB --l1d_size=32kB --l1i_assoc=4 --l1d_assoc=4 >>> --frame-capture --kernel=vmlinux.smp.mouse.arm >>> --disk=ARMv7a-Gingerbread.img --mem-size=256MB >>> >>> The restore command line adds "-r 1" >>> >>> The disk image is the Gingerbread disk image with bbench/busybox on >>> it, I just renamed it. I'm not passing a benchmark in, just booting >>> then checkpointing once the machine is up. After the checkpoint has >>> completed, the machine continues to run fine with any programs I >>> execute. I've since tested the same command line with gem5.opt, and >>> again it chokes. However, if I remove the caches and associated >>> options, the checkpoint runs correctly without error. >>> >>> >>> >>> On Sun, Jun 10, 2012 at 5:02 AM, Andreas Hansson >>> <[email protected]> wrote: >>>> >>>> Checkpointing is not part of the regressions so it could very well be >>>> that something broke it. >>>> >>>> Could you provide enough details to reproduce the problem? (e.g. the >>>> complete command line etc) >>>> >>>> Andreas >>>> ________________________________________ >>>> From: [email protected] [[email protected]] On Behalf >>>> Of Ira Ray Jenkins [[email protected]] >>>> Sent: Sunday, June 10, 2012 7:19 AM >>>> To: gem5 users mailing list >>>> Subject: [gem5-users] ARM Checkpoint SegFault >>>> >>>> When I checkpoint a machine, and attempt to restore from that >>>> checkpoint I get a SegFault. I'm not sure if this is a known issue or >>>> not. I'll try to attach the terminal output, the checkpoint is too >>>> large to attach. This has *only* happened recently, when I turned on >>>> caches with: "--caches --l1i_size=32kB --l1d_size=32kB --l1i_assoc=4 >>>> --l1d_assoc=4". >>>> >>>> -- IMPORTANT NOTICE: The contents of this email and any attachments are >>>> confidential and may also be privileged. If you are not the intended >>>> recipient, please notify the sender immediately and do not disclose the >>>> contents to any other person, use it for any purpose, or store or copy the >>>> information in any medium. Thank you. >>>> >>>> _______________________________________________ >>>> gem5-users mailing list >>>> [email protected] >>>> http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users >>> >>> _______________________________________________ >>> gem5-users mailing list >>> [email protected] >>> http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users >> >> >> _______________________________________________ >> gem5-users mailing list >> [email protected] >> http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users > _______________________________________________ > gem5-users mailing list > [email protected] > http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users > _______________________________________________ gem5-users mailing list [email protected] http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
